Our paper is highly cited because it describes a simple but useful computational tool for analyzing how charge is distributed in large electronic structure calculations. This so-called Bader partitioning of charge density has been used for many years, but the methods for doing the analysis are geared towards small molecular systems with well-defined bonding geometries.

“As computers get faster, and computational methods improve, we will one day be able to design materials from first principles calculations.”

Now, with the widespread use of plane-wave based density functional theory, the community needs more efficient computational tools to analyze large molecules and condensed phase materials. The algorithm described in this paper is faster and more robust than other methods for these large and complex systems. It is also implemented in a freely available software tool, which is gaining popularity in the field.

  Does it describe a new discovery, methodology, or synthesis of knowledge?

The paper describes an efficient and robust algorithm for partitioning electronic charge into Bader volumes. The algorithm is based upon a charge density grid, which can be generated from plane-wave based density functional theory calculations.

The algorithm in our paper allows computational researchers in the fields of chemistry and materials science to analyze properties of individual atoms more efficiently when they do calculations of large systems.

Before this paper, we were working on other methods for doing these kinds of calculations. When we had the idea behind the algorithm described in our paper, we realized that it was much simpler and more robust. The subsequent implementation of the method into a piece of software was not difficult. In fact, the biggest challenge we faced was getting the manuscript published. Our algorithm is quite simple, so it can appear obvious to people who have worked on related methods. After a year in the review process, we are happy to see it published and well received by the community.

The charge density population analysis we describe is useful for determining partial charges on atoms. There are, however, many other local properties that can be calculated based upon a Bader partitioning. In the future, we will extend our tools to calculate more properties of atoms in molecules and in materials systems, such as multipole moments, local densities of states, and local energies.

Density functional theory is a very powerful method for calculating the total energy of extended systems, but it will be even more useful if we can also determine local properties from the charge density.

This research, in itself, is not going to have any significant social or political implication. The field of computational material science, however, is going to have increasing implications, and tools like the one we have developed will help us use computers to compliment the experimental work being done to develop new materials and understand them at the microscopic scale. As computers get faster, and computational methods improve, we will one day be able to design materials from first-principles calculations.